Stochastic Modeling and Resource Dimensioning of Multi-Cellular Edge Intelligent Systems

TL;DR

This paper presents a stochastic framework for joint radio and computational resource dimensioning in multi-cell edge intelligent systems, optimizing deployment cost while ensuring quality of service under uncertainty.

Contribution

It introduces a unified probabilistic model combining network topology, interference, and workload profiling to optimize resource deployment with proven global optimality.

Findings

Smaller cells reduce transmission delay but increase per-request computing costs.

Larger cells decrease computing costs but increase transmission delay.

Densification benefits depend on frequency reuse scaling.

Abstract

Edge intelligence enables AI inference at the network edge, co-located with or near the radio access network, rather than in centralized clouds or on mobile devices. It targets low-latency, resource-constrained applications with large data volumes, requiring tight integration of wireless access and on-site computing. Yet system performance and cost-efficiency hinge on joint pre-deployment dimensioning of radio and computational resources, especially under spatial and temporal uncertainty. Prior work largely emphasizes run-time allocation or relies on simplified models that decouple radio and computing, missing end-to-end correlations in large-scale deployments. This paper introduces a unified stochastic framework to dimension multi-cell edge-intelligent systems. We model network topology with Poisson point processes, capturing random user and base-station locations, inter-cell interference, distance-based fractional power control, and peak-power constraints. By combining this with queueing theory and empirical AI inference workload profiling, we derive tractable expressions for end-to-end offloading delay. These enable a non-convex joint optimization that minimizes deployment cost under statistical QoS guarantees, expressed through strict tail-latency and inference-accuracy constraints. We prove the problem decomposes into convex subproblems, yielding global optimality. Numerical results in noise- and interference-limited regimes identify cost-efficient design regions and configurations that cause under-utilization or user unfairness. Smaller cells reduce transmission delay but raise per-request computing cost due to weaker server multiplexing, whereas larger cells show the opposite trend. Densification reduces computational costs only when frequency reuse scales with base-station density; otherwise, sparser deployments improve fairness and efficiency in interference-limited settings.